PT2054443E - Antibodies to an epitope of agr2, assays and hybridomas - Google Patents

Abstract

Provided is a monoclonal antibody, or an antigen binding fragment thereof, which binds specifically to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2. Such monoclonal antibodies are of prognostic and diagnostic utility in the investigation of cancer, particularly metastatic cancer. The antibodies described may also be used in prognosis or diagnosis of inflammatory diseases. Also provided are kits and solid supports comprising such antibodies, as well as the therapeutic use of antibodies of the invention.

Description

1

DESCRIPTION

" ANTIBODIES AGAINST AN EPITHOPE OF AGR2, TESTS AND HYBRIDOMES " The present invention relates to monoclonal antibodies and to antigen binding fragments of such monoclonal antibodies. The invention also relates to hybridomas capable of producing such antibodies. Other aspects of the invention also relate to uses of the disclosed monoclonal antibodies in diagnostic and prognostic assays of metastatic disease. The above gradient 2 protein (AGR2) is the human homolog of the gene specific for the Xenopus cement gland (XAG-2). The amino acid sequence of AGR2 is shown in Sequence ID No. 1. Increased AGR2 expression has been shown to correlate with estrogen receptor (ER) positivity in breast carcinoma specimens. Induced expression of AGR2 appears to give rise to a metastatic phenotype in what would otherwise be benign non-metastatic cells. Cancer is the second most common cause of death in the Western world, and is the leading cause of death among those under 85 years of age. One in four of the population will die of cancer, and these deaths usually occur as a consequence of untreatable metastatic disease. The spread of metastases through the body increases the number of tissues damaged by cancer, while making cancer treatment increasingly difficult. The number of metastases that can develop from a single primary tumor will generally be too large to be treated, and the number of different sites requiring treatment will be too abundant and varied. The ability of metastatic cancers to " reappear " or " appearing " even after excision or resolution of a primary tumor and systemic treatment, contributes greatly to the difficulties encountered in cancer treatment and also to the distress experienced by cancer patients in the face of uncertainty if a treatment was effective to render them free of cancer.

It will be understood, in light of the negative influence of cancer on the population as a whole, that there is a well-recognized need for methods that can be used to diagnose metastatic (metastatic) cancer. Many existing methods lack sensitivity and / or reliability and, therefore, the results of these methods can not be seen as conclusive. A reduced sensitivity of the diagnostic assays may lead to incorrect diagnoses based on the incorrect test results.

There is also a need for assays that can be used in prognosis to assess the likelihood of a patient to develop metastatic cancer. The limitations of sensitivity and reliability of existing assays for metastatic disease also mean that certain patients with an increased risk for developing metastatic disease are not necessarily identified using existing assays. Failure to identify such patients may mean that opportunities for therapeutic intervention are missed before the onset of symptoms of metastatic disease. 3

There is also a need for prognostic assays that are capable of indicating whether a patient will respond favorably or not to treatment for metastatic cancer. In general, treatments for metastatic cancer comprise relatively " hard " regimens. Such treatment regimens and the discomfort effects they may cause are generally justified in the case of patients for whom they are not expected to respond to less stringent treatment. However, it will generally be preferred to avoid subjecting patients who are likely to respond favorably to a milder " to unnecessary harsh regimes. Fletcher et al. (2003) disclose a polyclonal antibody against an association of two peptides representing two adjacent fragments of hAG-2 with minimal homology to other known members of the family. It is an object of certain embodiments of the invention to overcome or alleviate problems associated with the prior art. For example, it is an object of certain embodiments of the present invention to provide diagnostic assays for metastatic disease that are more sensitive than those provided by the prior art. It is an object of certain embodiments of the present invention to provide diagnostic assays for metastatic disease that are more reliable than those provided by the prior art. It is an object of certain embodiments of the invention to provide prognostic assays for metastatic disease that are more sensitive than those provided by the prior art. It is an object of certain embodiments of the invention to provide prognostic assays for metastatic disease that are more reliable than those provided by the prior art. It is an object of certain embodiments of the invention to provide agents that can be used in diagnostic and / or prognostic assays for metastatic disease with greater sensitivity and / or reliability than those provided by the prior art.

In a first aspect, the invention provides a monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2. Sequence ID No. 2 corresponds to amino acid residues 26 ° to 40 ° of human AGR2. The inventors have found that a monoclonal antibody, or antigen-binding fragment thereof, having the specificity defined above provides marked advantages in terms of use compared to the AGR2 antibodies of the prior art. In particular, the antibodies of the invention may be used in assays for the prognosis or diagnosis of cancer, and in particular metastatic disease, which have greater sensitivity than similar assays provided by the prior art. Although AGR2 expression is strongly associated with mammary cancers, the expression of this marker is also present in a wide range of cancers including esophageal squamous cell carcinoma, pancreatic carcinoma and prostate carcinomas. Accordingly, the inventors believe that the antibodies and methods of the invention may be used with respect to cancer and metastatic disease, associated with any form of primary cancer. " Antibodies of the invention, " as referred to in the present disclosure, may be any monoclonal antibody having specificity for an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2. The antibody produced by the hybridoma PZ7A10F10 represents a particularly preferred example of an antibody of the invention. PZ7A10F10 was deposited under the Budapest Treaty in the European Collection of Cell Cultures (ECACC), Porton Down, Salisbury, on 22 August 2006, and was given Accession Number 06082201. The monoclonal antibody produced by this hybridoma may be referred to, for purposes of the present description, as " anti-AGR2 / PZ7A10F10 ". The anti-AGR2 / PZ7A10F10, or the antigen-binding fragments thereof, represents a preferred antibody, or preferred antigen-binding fragments, for use in all aspects and embodiments of the invention. AGR2 / PZ7A10F10 is a K light chain IgG antibody.

As used in the present description, the term antibody of the invention should also be understood, unless the context otherwise requires, as encompassing antigen binding fragments of antibodies of the invention, as well as antigen binding derivatives of such antibodies or fragments of antibody.

Any / any antigen-binding fragment (s) of the antibodies of the invention may be prepared according to techniques known to those skilled in the art. The smallest functional antigen-binding fragments of antibodies of the invention may comprise the heavy (VH) or light (VL) chain variable regions of such antibodies. These fragments may have a molecular mass of approximately 13 kDa, or less than one tenth the typical size of a complete antibody.

Antigen-binding fragments of this type can be well expressed in bacterial, yeast and mammalian cell systems. Such fragments may also be resistant to conditions that would otherwise be deleterious, such as lyophilization or thermal denaturation.

Antibodies of the invention may have Variable and Constant domains. It will be understood that antigen-binding fragments (e.g., scFV antibodies) comprising essentially the variable region of an antibody lacking any constant regions are also encompassed by the present invention.

The antibodies of the invention may be " humanized " using techniques that will be well known to those skilled in the art. Humanization may be at least partially accomplished by engineered antibodies that utilize V region sequences of non-human mAbs (e.g., rodents) and C-region (and ideally V-region FR) sequences of human antibodies. The resulting 'engineered' mAbs are less immunogenic in humans than the rodent mAbs from which they are derived and are thus more suitable for clinical use, and are thus more susceptible to use in in vivo assays. Further humanization of antibodies may involve CDR grafting or antibody remodeling. Such antibodies are produced by transplating the heavy and light chain CDRs of a rodent mAb (which forms the antigen-binding site of the antibody) at the corresponding framework regions of a human antibody.

It will be understood that in at least some extent the increased sensitivity of the monoclonal antibodies of the invention arises as a consequence of the specificity of their binding to AGR2. Accordingly, it may be preferred that the monoclonal antibodies of the invention specifically bind to AGR2,7 but not to similar or related proteins. It may be particularly preferred that the monoclonal antibody of the invention is one which does not bind to the above gradient 3 (AGR3) protein. Indeed, this feature is so advantageous that the invention also provides a monoclonal antibody with specificity for AGR2, while lacking any specificity for AGR3.

Antibodies of the invention are particularly well suited for use in diagnostic and / or prognostic applications. Accordingly, in a third aspect, the invention provides a method of diagnosing cancer, and in particular metastatic disease, using an antibody of the invention. In a fourth aspect, the invention provides a method of evaluating the likelihood of developing cancer, and in particular metastatic disease, using an antibody of the invention. For the purposes of the present description a " method of the invention " should be understood as encompassing any use of the antibodies of the invention in diagnosis or prognosis. Unless the context otherwise requires any method of the invention may be practiced using any antibody of the invention. The use of the antibodies of the invention in diagnosis is particularly advantageous since the diagnostic sensitivity of the assays using the antibodies of the invention is greatly increased over that which can be achieved using techniques of the prior art. While not wishing to be bound by any hypothesis, the inventors believe that this increased sensitivity of the diagnostic assays may stem from the increased sensitivity of the antibodies of the invention.

The inventors have found that the antibodies of the invention can be used in prognostic or diagnostic assays in which the ability of the antibodies of the invention to assess tumor cells present in a sample of the patient is evaluated. The inventors have found that any binding of the antibodies of the invention to tumor cells present in a sample of the patient can be considered as indicative, in terms of prognosis or diagnosis, of metastatic disease. In particular, binding of antibodies of the invention to approximately 1% or more of tumor cells present in a patient sample is extremely prognostic and / or diagnostic of metastatic disease.

Accordingly, an assay according to the third or fourth aspects of the invention may comprise contacting a patient sample with a monoclonal antibody, or an antigen-binding fragment thereof, according to the invention; and assessing the binding of the antibody to the patient sample, wherein binding of the antibody to the patient sample is diagnostic and / or prognosis of cancer, and in particular of metastatic disease, in the patient. Preferably, the patient sample may be a sample containing tumor cells. In this case the binding of the antibody of the invention (or antigen-binding fragment thereof) to the tumor cells can be assessed. In case the antibody of the invention binds to tumor cells present in the patient sample, i.e. is particularly diagnostic and / or prognostic of metastatic disease in the patient.

An assessment can be made as to the proportion of tumor cells in a sample of the patient that are bound by an antibody of the invention (or an antigen-binding fragment thereof). For example, a prognostic and / or diagnostic result indicative of metastatic disease in patient 9 may be indicated by binding of the antibody (or antigen-binding fragment thereof) to 1% or more of the tumor cells present in the patient sample.

Higher incidences of binding of the antibodies of the invention to tumor cells in a sample of the patient are associated with a reduction in the life expectancy of the patient from which the sample is derived. In the event that an antibody of the invention binds to approximately 5% or more tumor cells in a sample of the patient, this may indicate that the patient would receive rapid intervention to try to treat the cancer and possible metastases. Assays of this type can also be used to determine suitable therapeutic regimes for the treatment of cancer, representing those with relatively high binding affinity of the antibodies of the invention (up to about 5% or more of tumor cells present in a patient sample) suitable candidates for treatment with regimes that may be relatively harsh if such regimes provide benefits administered quickly.

The diagnostic assays of the invention may be advantageously used to identify individuals suffering from cancer, and in particular metastatic disease. Diagnostic assays according to the third aspect of the invention may comprise contacting a patient sample with an antibody of the invention, and evaluating the binding of the antibody to the patient sample, wherein binding of the antibody to the patient sample is diagnosis of cancer, and in particular of metastatic disease, in the patient. Binding of the antibody to the patient sample can be assessed by any relevant technique known to those skilled in the art, and a suitable technique may be selected by the skilled artisan having regard to the nature of any reporter unit that may be attached to the labeled antibody. Suitable reporter units that can be used as tags attached to the antibodies of the invention are considered elsewhere in the specification.

It will be appreciated that the increased sensitivity of the diagnostic assays disclosed herein has a number of advantages. The most important of these advantages is that the diagnostic assays of the invention may allow the user a greater degree of confidence in the diagnostic result than in results provided by prior art assays. The increased sensitivity of the assays of the invention allows the detection of AGR2 in patient sample at lower levels than has previously been the case. This allows for earlier diagnoses of cancer, and in particular of metastatic disease, than has been possible using the methods of the prior art, and thus allows treatment to begin earlier after the onset of the disease. Early treatment is important to achieve maximum therapeutic efficacy and thereby increase a patient's chances of survival. Diagnostic detection of AGR2 using the assays of the invention also allows the selection of appropriate therapeutic regimens for the treatment of cancer, and in particular of metastatic disease, which affects the patient. In particular, the inventors' observations suggest that patients in whom cancer (in particular metastatic disease) causes AGR2 expression may be subject to shorter survival than equivalent patients who do not express AGR2. This effect seems particularly noteworthy in the case of patients with estrogen receptor (ER +) treated with tamoxifen. This observation indicates that AGR2 expression, which can be determined with greater certainty and sensitivity using the assays of the invention, is indicative of a form of metastatic disease that is apparently less severe, and characterized by ER + that do not respond well to therapy hormonal. Therefore, diagnosis using the assays of the invention may lead to a clinician with responsibility for the diagnosed patient to avoid hormone therapy, and instead to choose alternative therapies that are likely to have greater beneficial results.

Finally, the increased sensitivity conferred by the assays of the invention means that an observation that AGR2 is not being expressed can be more reliably accepted as evidence of the absence of metastatic disease. Thus, patients who are diagnosed as having no metastatic disease (or no longer present after successful therapy) may have greater confidence that this observation is indeed correct.

In addition to the diagnostic uses described above, the monoclonal antibodies of the invention are also amenable to use in prognostic assays that may be used to assess the risk of the patient having an adverse clinical outcome due to the development of cancer, and in particular metastatic cancer. A study by Innes et al. (2006) indicates that ER + patients exhibiting AGR2 expression tend to suffer from reduced survival compared to similar patients in whom AGR2 is absent. Said study discloses a polyclonal antibody specific for AGR2 and not for AGR3.

In particular, the monoclonal antibodies of the invention may be used in prognostic assays which provide an indication of the degree of satisfaction with which a patient with cancer, and in particular metastatic cancer, will respond to the treatment of the disease. The prognostic assays according to the fourth aspect of the invention may comprise contacting a patient sample with an antibody of the invention, and analyzing for binding of the antibody to the patient sample, wherein binding of the antibody to the sample of the antibody indicates that the patient is unlikely to respond favorably to the treatment of metastatic disease. Such patients may benefit from treatment using strict therapeutic regimens. Details of suitable stringent treatment regimens that may be used to treat metastatic cancer may be determined relative to the specific tissues or organs affected by the metastatic disease. Appropriate criteria for selection of suitable regimens will be apparent to one skilled in the art and may for example be selected by a clinician in charge of the patient.

The inventors believe that the assays of the invention can be used prognostically since detectable expression of AGR2 may occur prior to the development of clinically detectable cancer, and in particular metastatic disease. In addition, the inventors believe that the expression of AGR2 by metastases could affect the behavior or development of metastases. In addition, AGR2 may be detectable in patient samples prior to the development of cancer, and in particular of clinically recognizable metastatic disease. Thus, the prognostic assays of the invention may permit the identification of the presence of AGR2, and thus a patient is more likely to have shorter survival as a consequence of cancer or metastatic disease. Recognition of a patient's increased likelihood for reduced survival resulting from metastatic disease may allow prophylactic intervention (to prevent further development of the disease) and / or further monitoring of the development of metastatic disease to allow treatment to be initiated sooner the onset of disease.

The monoclonal antibodies of the present invention may be used in prognostic assays or diagnostic assays as described elsewhere in the specification. The inventors believe that assays using the monoclonal antibodies of the invention may be used in particular for the diagnostic or prognostic evaluation of cancers selected from the group consisting of: prostate cancer; ovarian cancer; and pancreatic cancer. It may be preferred that the evaluations of these selected cancers be performed using assays of the invention, as described elsewhere in the specification.

The inventors have surprisingly found that the monoclonal antibodies of the invention can also be used in the prognosis or diagnosis of diseases other than cancer. In particular, the inventors believe that the monoclonal antibodies of the invention can be used for the prognostic or diagnostic evaluation of inflammatory diseases. Rheumatoid arthritis represents a particularly preferred example of such an inflammatory disease which can be evaluated in terms of prognosis or diagnosis using monoclonal antibodies in accordance with the present invention.

A prognostic or diagnostic assay for an inflammatory disease may comprise obtaining a sample from the patient; contacting the patient sample with a monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2); and analyzing in connection with the binding of the antibody to the patient sample, wherein binding of the antibody to the patient sample is a prognosis or diagnosis of an inflammatory disease in the patient.

Any patient sample that may contain AGR2 protein may be used in the prognostic or diagnostic assays of the invention (either with regard to cancer and metastatic disease or inflammatory diseases). These may include essentially acellular samples, such as urine, cerebrospinal fluid or lymph, into which AGR2 may be deposited. Preferably, a suitable sample may be a sample comprising cells capable of expressing AGR2. Examples of such samples include blood samples or samples (such as biopsy samples) taken from other tissues. The inventors have found that histological samples or cryotomy samples may be of particular utility in the prognostic or diagnostic assays of the invention.

Antibodies and assays of the invention may be used in prognosis or diagnosis for many forms of cancer. Particular forms of cancer that may be advantageously investigated using the antibodies or assays of the invention include metastatic cancer; prostate cancer; ovarian cancer; and pancreatic cancer; breast cancer; stomach cancer; cancer of the esophagus; and colon cancer. It will be understood that the use of the antibodies and assays of the invention in diagnostic may be particularly useful in the case of cancers that may be difficult to otherwise diagnose, including cancers in " interior sites " of the body (i.e., non-palpable from outside the body), such as pancreatic cancer or ovarian cancer; cancers that otherwise may remain " asymptomatic " for prolonged periods, such as stomach cancer; or cancers in which AGR2 appears to be highly expressed; like cancers of the colon. 15

It will generally be preferred that the assays of the invention, whether prognostic or diagnostic, are used with respect to human patient samples.

The assays of the invention may employ any means suitable for detecting binding of an antibody to its antigen which are known to those skilled in the art. Suitable methods may be selected in relation to the nature of any reporter unit used to label the antibodies of the invention. Suitable techniques include, but are not limited in any way to, flow cytometry (such as fluorescence-activated flow cytometry FACS) and enzyme-linked immunosorbent assays (ELISAs), and assays using nanoparticles. It is particularly preferred that an assay of the invention is one involving immunocytochemistry in which the tumor cells are exposed to an antibody of the invention, and the level of labeling of the cells evaluated. Greater marking of cells will generally be indicative of poor clinical outcome. An adequate statistical analysis of the results of such assays may be carried out according to the tests outlined in Example 2 of the Experimental Results section. The invention also provides, in a fifth aspect, a hybridoma capable of producing an antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2.

A " hybridoma of the invention " may be any hybridoma produced according to the methods of the invention.

Preferably, a hybridoma of the invention will be a hybridoma capable of producing a monoclonal antibody of the invention. The hybridoma deposited as PZ7A10F10 under accession number 06082201 represents a particularly preferred example of a hybridoma of the invention.

A hybridoma producing the monoclonal antibody of the invention can be prepared by applying a routine immunization procedure using the peptide fragment KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2 as an antigen of sensitization, followed by a cell fusion procedure using a routine cell fusion technique, and a cloning procedure using a routine cloning technique.

In a suitable example, a non-human animal, such as a rat, rat, rabbit, guinea pig, pig, sheep, goat or chicken may be the animal to be immunized. Preferably, the animal to be immunized is a rat or rat, and most preferably a mouse. Because myeloma cells used as homologous cells for the fusion of cells are generally derived from mice, it may be particularly preferred to immunize mice when selecting the source of antibody producing cells. The peptide fragment KPGAKKDTKDSRPKL (Sequence ID No. 2), corresponding to amino acid residues 26 to 40 of the full-length AGR2, is used as the sensitizing antigen. This antigen can be administered by any suitable route (for example, intraperitoneally, subcutaneously or in the paw plant) to generate an immunological response in the immunized animal. Preferably, the immunizing antigen may be administered by subcutaneous injection. The skilled person will appreciate that the antibodies of the invention can be generated using AGR2 fragments comprising SEQ ID NO: 2 (including full length AGR2) as an immunogen, followed by selection of the hybridomas which produce an antibody that binds to an epitope within KPGAKKDTKDSRPKL (Sequence ID No. 2).

Antibodies of the invention may be generated using AGR2 from non-human species (or corresponding sequences of equivalent proteins from non-human species) as long as antibodies are produced having the necessary binding specificity. Immunization to produce antibodies of the invention may be performed using any suitable method known to the skilled person. For example, a suitable method may comprise administering a suitable dose of KPGAKKDTKDSRPKL (Sequence ID No. 2) diluted and suspended in a vehicle such as phosphate buffered saline, saline or the like to recipient animals on a weekly basis for one to three months.

Upon completion of a suitable immunization regimen, spleen cells, lymphocytes, peripheral blood cells or other antibody producing cells may be harvested from the immunized animals. In general, spleen cells, excised after the last administration of the sensitizer antigen as the antibody producing cells, may be preferentially used. In either case, the antibody-producing cells harvested may then be subjected to fusion of cells with particular strains of myeloma cells, a tumor cell line, to prepare hybridoma cells.

A range of suitable myeloma cell lines that can be used in the production of hybridoma cells will be known to those skilled in the art. For example, cell lines, such as P3-NSI / 1-Ag4-1 cells (short for NS-1 cells) (Kohler et al., Eur. J. Immunol., 6: 511-18 (1976)), SP2 / 0-Agl4 cells (SP2 cells) (Schulman et al., Nature, 276: 269 (1978)) and FO cells (DeSaint Groth et al., J. Immunol. Meth., 35: 1 (1980) ) are all suitable to be used to produce hybridomas according to the present invention. It may be preferred to use SP2 cells. In order to facilitate recovery of the antibody of interest from a supernatant produced by the hybridoma culture, it may be preferable to use a myeloma cell line which does not secrete the immunoglobulin inherent in the myeloma cells. For example, it may be preferred to use NS-1 cells as myeloma cells for the fusion of cells.

The methods by which an antibody producing cell and a myeloma cell can be fused are now a matter of routine procedure for those skilled in the art. Merely by way of example, cell fusion can be achieved according to the protocol used by Kohler and Milstein (Kohler et al., 1975, Nature, 256, 495). The cell fusion procedure can be performed in a standard nutrient medium in the presence of a fusion promoter. The suitable fusion promoter that can be used to produce hybridomas according to the present invention includes polyethylene glycol (PEG) or Sendai virus. The suitable PEG may have a number average molecular mass of 1000 to 6000. Fusion of cells is performed using antibody-producing cells and myeloma cells mixed in a predetermined ratio. Suitable ratios may range from about 1: 1 to about 10: 1. RPMI 1640 can be used as a culture medium in which the fusion of cells can occur, since this medium favors the growth of myeloma cells. In order to effect cell fusion, the two cell types (antibody producing cells and myelomas) can be mixed in RPMI 1640 and maintained under normal cell culture conditions. A solution of polyethylene glycol in a concentration of 30 to 60% (m / v) can then be added to initiate the melting of cells. Finally, once the cell fusion has occurred, an additional volume of a suitable medium can be added and the produced hybridoma recovered by centrifugation to remove the supernatant. The recovered hybridoma can be selected by culturing in a standard selection medium such as HAT medium containing hypoxanthine (H), aminopterin (A) and thymidine (T). Culture in HAT medium can be continued for a period of several days to a few weeks until all cells other than the hybridoma have been killed.

When the presence of hybridoma colonies is confirmed, the cultured supernatant can be screened for the presence of the monoclonal antibody. Suitable screening for the antibody in the culture supernatant can be, for example, performed by analyzing antibody activity in the culture supernatant by an ELISA technique using immobilized cells such as the antigen (Ando Tamie et al., Introduction to Monoclonal Antibody Experiments Protocols , J. Dansha Scientific (1993), 126). It will be understood that cells suitable for use in such a technique should be those which express AGR2, such as metastatic cancer cells. Once the presence of the hybridoma and the desired monoclonal antibody is confirmed, the hybridoma can be cloned for subsequent propagation using well known techniques (for example ring cloning or cloning by limiting dilution). 20

Following the methods described above allows the production, isolation and cloning of a hybridoma producing the monoclonal antibody of the invention. Suitable techniques for the maintenance of hybridomas are well known to those skilled in the art. For example, hybridomas may be cultured and subcultured using known cell culture media, such as RPMI 1640 or modified Dulbecco's essential medium (DMEM). Samples of hybridomas produced in the manner described above and producing the monoclonal antibody of the invention may be stored in liquid nitrogen before being thawed for future use.

The hybridomas of the invention are suitable for cultivation on a large scale, to facilitate the production of large amounts of the antibodies of the invention. Hybridomas of the invention may, for example, be cultured in 15% fetal calf serum (FCS) -RPMI 1640 and the monoclonal antibodies of the invention may be prepared from the culture supernatant.

As an alternative, the hybridomas of the invention can be injected intraperitoneally into experimental mice to form ascitic tumors. The monoclonal antibodies of the invention can then be prepared from the ascitic liquid.

The monoclonal antibodies of the invention, in any form, can be purified using well known antibody purification techniques. Suitable examples of antibody purification technology that can be used to purify the antibodies of the invention comprise precipitation (salt addition) with ammonium sulfate or the like, ion exchange chromatography using a diethylamino ester derivative 21 (DEAR) , a carboxymethyl derivative (CM), or the like, hydroxyapatite chromatography, gel filtration chromatography, and affinity chromatography using Protein A or Protein G, among others, including binding to the antigen against which the antibody was created. It will be understood that combinations of the techniques suggested above may be used in the purification of the monoclonal antibody. The invention also provides a method for the preparation of a monoclonal antibody of the invention, the method comprising immunizing an animal with the peptide KPGAKKDTKDSRPKL (Sequence ID No. 2) and then screening and isolating the antibody. The invention also provides a method for the preparation of a hybridoma which produces an antibody of the invention, the method comprising immunizing an animal with the peptide KPGAKKDTKDSRPKL (Sequence ID No. 2), obtaining an antibody producing cell of the immunized animal ; and fusion of the antibody-producing cell with a myeloma cell. The skilled artisan will readily understand that the monoclonal antibodies of the invention can be used to generate antigen-binding antibody fragments. These can be generated by manipulating the existing monoclonal antibodies or using the monoclonal antibodies as a template for the production of suitable fragments. The mechanisms by which such fragments can be produced will be well known to those skilled in the art.

Merely by way of example, Fab, F (ab ') 2 and other immunoreactive fragments can be obtained by digesting the monoclonal antibody of the invention with a proteolytic enzyme that does not decompose the antigen binding site (Fab), such as papain , pepsin, or the like. The antigen-binding fragments thus generated can then be isolated and purified using routine techniques. Antigen binding fragments with the specificity of the monoclonal antibodies of the invention can be used in the same manner as the monoclonal antibodies themselves.

The monoclonal antibodies or antigen-binding fragments according to the invention may be used in a number of assays (including, but not limited to diagnostic and prognostic assays of the invention), in which it may be beneficial to be able to obtain information about the location or binding of the antibody or fragment. In such cases it may be preferred that the antibody, or antigen-binding fragment, is labeled using a reporter unit. Such reporter units may be directly or indirectly linked to an antibody of the invention.

Suitable reporter units will be well known to those skilled in the art, as will methods by which they may be attached to the antibodies of the invention. A suitable reporter unit may be selected from the group consisting of a fluorescent moiety; a luminescent unit; a bioluminescent unit; a radioactive material; a prosthetic group; a colorimetric group; nanoparticles with suitable detectable properties and a chromogenic unit.

Merely by way of example, and without restriction, suitable fluorescent units may be selected from the group consisting of: fluorescein isothiocyanate (FITC); rhodamine (TRITC); phycoerythrin; allophycocyanin; coumarin (AMCA); Texas red; and cyanine (Cy 2, Cy 3 or Cy 5). 23

Other suitable fluorescent units that may be used in the labeling of antibodies of the invention will be readily apparent to the skilled person.

A suitable labeling of antibodies of the invention with FITC can be achieved using the following protocol. The antibody of the invention is prepared as a 2 mg / ml solution in 0.1M sodium carbonate (pH 9.0). The FITC is dissolved in 1 mg / mL DMSO and slowly added to the antibody solution (at a concentration of about 50pg FITC per mL of antibody). The antibody and FITC mixture is incubated in the dark for 8 hours at 4 degrees. Upon completion of this first incubation, 50 mM NH4 Cl is added and the resulting mixture is incubated for 2 hours at 4 degrees. Finally, xylene cilanol and 0.1% and 5% glycerol respectively are added. The antibody of the labeled invention can then be separated by gel filtration.

For purposes of the present disclosure, luminescent and bioluminescent units may be considered to encompass units having their own luminescent properties and units capable of giving luminescent products. Luminol represents a suitable example of a luminescent unit which can be used to label antibodies of the invention. Luciferase provides an example of a bioluminescent unit (in this case a unit capable of generating a luminescent product) which can be used to label antibodies of the invention.

Suitable radioactive materials that can be used to label antibodies of the invention will be apparent to the skilled person. Merely by way of illustrative example, suitable radioactive materials may include radioisotopes selected from the group consisting of: 125 I; 131I; 35S; 3H; 14C; 32P; 99mTC and niIn. 24

An antibody of the invention may be labeled using prosthetic groups which bind to one another with high specificity. An example of such a suitable prosthetic group that will be well known to those skilled in the art is biotin. An antibody of the invention may be labeled with biotin and the biotinylated antibody exposed to a binding partner specific for biotin (for example avidin or streptavidin). The binding partner may carry a separate label, such as a fluorescent label.

Colorimetric units that can be used to label antibodies of the invention include, but are not limited to: colloidal gold; and colored glass or plastic beads (e.g., polystyrene, polypropylene, latex or the like).

Examples of chromogenic units that can be used to label antibodies of the invention include chromogenic enzymes such as horseradish peroxidase and alkaline phosphatase. Various substrates that can be used to generate detectable products representative of the activity of these enzymes are well known and are in fact commercially available from many suppliers of laboratory reagents.

In addition to the direct labeling of antibodies of the invention, in which the reporter units can be attached directly to the antibody or the antigen-binding fragment, the antibodies of the invention may also be labeled using indirect labeling techniques. Methods by which antibodies, like the antibodies of the invention, can be labeled indirectly are well known to those of ordinary skill in the art. 25

A suitable way by which the antibodies of the invention can be indirectly labeled is by a strategy of " primary antibody " " secondary antibody ". Briefly, in such a strategy the antibody of the unlabelled invention is used as a " primary antibody " capable of binding to AGR2 in a sample (e.g., a patient sample). &Quot; secondary antibody " (chosen to react exclusively with the antibody of the invention and not with other materials in the sample) is then used to bind the primary antibody. Thus, the antibody of the unlabeled invention is effectively attached to the tag attached to the secondary antibody. The labeling of antibodies of the invention (either directly or indirectly) allows the detection of these antibodies. Typically the unbound molecules bearing the chosen label (such as directly labeled antibodies or unbound secondary antibodies) will be removed from the sample whereby essentially only the label associated with bound antibodies of the invention remains. Detection of the tag is then considered to represent the detection of bound antibodies of the invention. The means by which the label will be detected will depend on the nature of the label selected. For example, a fluorescent label will be detected by illuminating the sample (containing the antibody of the invention) bound with light at the excitation wavelength of the fluorescent label and detecting the presence of light at the emission wavelength of the selected fluorophore.

A labeled antibody with a chromogenic enzyme can be detected by incubating the sample (containing the antibody of the invention) with a selected chromogenic substrate of the enzyme 26 and detecting the presence of the stained product as a result of the action of the enzyme on the substrate.

The ways in which antibodies of the invention that have been labeled using alternative reporter units can be detected will be well known to those skilled in the art.

In other embodiments, the invention also provides kits for use in diagnostic or prognostic assays of the invention, the kits comprising a monoclonal antibody or antigen-binding fragment thereof of the invention. It will be understood that such kits may be used in the prognosis or diagnosis of one or more diseases selected from the group consisting of: cancer; (and in particular metastatic cancer, prostate cancer, ovarian cancer, pancreatic cancer, breast cancer, stomach cancer, esophageal cancer, or colon cancer); an inflammatory disease in general; and rheumatoid arthritis in particular. The kits according to the present invention may also include additional components. Such components may include, for example, at least one item selected from the group consisting of: instructional materials (in any form, including printed or computer readable materials); reagents for use in detecting antibody binding (including reagents, as labeled secondary antibodies, which may be used in the visualization of the bound antibodies of the invention); reagents for use in incubating the antibody (as solutions which may be used for dilution of primary or secondary antibodies, or reagents for use in chromogenic labeling protocols); and agents for the visualization of cell nuclei. Preferred labeled secondary antibodies include fluorescently labeled antibodies and labeled antibodies with a chromogenic agent. The antibody of the invention may be provided as a panel of reagents (such as suitable antibodies other than those of the invention) that have prognostic or diagnostic utility to be included in a kit according to the invention. The invention also provides a solid substrate to which is attached a monoclonal antibody, or antigen-binding fragment, of the invention. For example, the substrate may comprise a matrix comprising antibodies of the invention. Suitable substrates may also include microspheres. It will be understood that the antibodies of the invention may be provided on a solid substrate as a panel of reagents (including antibodies other than those of the invention) that have utility in prognosis or diagnosis.

The inventors believe that the antibodies of the invention may be capable of blocking the AGR2 function expressed by tumor cells. Given the exceptionally high specificity of the antibodies of the invention (as illustrated by their greater utility in prognosis and diagnosis compared to the previously disclosed antibodies), these antibodies may represent extremely promising therapeutic agents, since they can target AGR2 without blocking the function related components that are not involved in the progression of cancer. Accordingly, the invention provides an antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2 to be used as a medicament. The antibody, or antigen-binding fragment, to be used in this aspect of the invention may preferably be a monoclonal antibody produced by the hybridoma PZ7A10F10. 28

Drugs manufactured using the antibodies of the invention may be useful in the treatment of cancer and in the prevention and / or treatment of metastatic disease (in which the ability of antibodies to block AGR2 function may help prevent the proliferation and spread of metastases). Suitable modifications of the antibodies of the invention that will be used as therapeutic agents will be apparent to those skilled in the art. Merely by way of example, it may be preferred to use fragments of such antibodies if it is desired to achieve intracellular activity leading to a therapeutic effect. Additionally or alternatively, it may be intended " humanize " the antibodies (as contemplated elsewhere in the specification) to reduce the likelihood of undesired side effects caused by the patient's immune response to the exogenous antibody.

Antibodies of the invention, or antigen-binding fragments of such antibodies, may also be used in the prevention and / or treatment of inflammatory diseases. The amino acid sequence of AGR2 (Sequence ID No. 1) is illustrated in the Sequence Information section, along with the amino acid sequence of the sensitizer antigen SEQ ID NO: 2, a peptide fragment of AGR2. Amino acid residues 1-20 of Sequence ID No. 1 constitute a secretory signal sequence. The peptide fragment shown in Sequence ID No. 2 comprises amino acids 26-40 of Sequence ID No. 1. The invention will now be further described with respect to the following Experimental Results and Figures in which:

Figure 1 shows the western blot results to determine the specificity of the anti-AGR2 monoclonal antibody produced according to Example 1. Lane 1 shown in Figure 2 was loaded with 0.5 pg AGR2 while lane 2 was loaded with , 5æg of AGR3. Antibody binding (detected by chemiluminescence) indicates that the monoclonal antibody binds strongly to AGR2 (shown by the strong signal in lane 1) but does not bind to AGR3 (indicated by the lack of signal in lane 3).

Figure 2 shows immunocytochemistry of patient primary tumor samples using a monoclonal antibody of the invention to label AGR2. Panel 2A: normal breast; Panel 2B: a section of tumor showing negative development for AGR2; Panel 2C: a section of tumor showing positive development of carcinoma cells; 2D Panel: a higher 2C amplification that shows a strong immunocytochemical development for AGR2. The arrow indicates the tumor cells that gave positive when developed in relation to AGR2 using the antibody of the invention. Magnification A-C 40x; D 12 5x.

Figure 3 shows the frequency of the development category (according to Table 1) of different carcinoma samples using an antibody according to the invention. Primary carcinomas of 320 patients were immunocytochemically revealed with respect to AGR2 with the monoclonal antibody of the invention.

Figure 4 shows survival traffics for each development category using the antibody of the invention. The cumulative survival rate represents the percent survival of patients in each disclosure category every 12 months over the study period 30 of 240 months. A total of 315 patients were evaluated. (a) patients with negative development relative to AGR2 (-), 100% for 107 patients; (b, -----) borderline disclosure patients (+/-), 100% for 93 patients; (c, - - -) patients with 5-25% development (+), 100% for 61 patients; (d, - -) patients with 25-50% of development (++), 100% for 26 patients and (and, -) patients with 50% or more of development (+++ / ++++ ), 100% for 28 patients. There were 103 censored observations in a (19 deaths from other causes); 39 censored observations in b (16 deaths from other causes); 17 censored observations in c (10 deaths from other causes); 4 censored observations in d (4 deaths from other causes) and 10 censored observations in e (9 deaths from other causes). Overall the five graphs are very significantly different (Wilcoxon χ2 statistic = 103.157, 4 degrees of freedom, P < 0.0001).

Figure 5 shows survival charts for the negative (-), border (+ / -) and all positive (+, ++, +++ / ++++) groups relative to AGR2. The cumulative proportion of surviving patients represents the percent survival of patients in each disclosure category every 12 months over the study period of 240 months. A total of 315 patients (a, -), patients with negative development with respect to AGR2 (-), 100% for 107 patients; (b, - -), patients with borderline disclosure (+/-), 100% for 93 patients; (c, ----), patients with positive (+, ++, ++++ / ++++), 100% for 115 patients. There were 103 censored observations in a (19 deaths from other causes); 39 censored observations in b (16 deaths from other causes) and 31 censored observations in c (23 deaths from other causes). The three graphs were very significantly different (Wilcoxon χ2 statistic = 101.49, 2 degrees of freedom, P < 0.0001). 31

Figure 6 shows survival charts for negative (-) and all other positive (+ / -, +, ++, +++ / ++++) groups using 1% threshold level for the disclosure of AGR2. The cumulative proportion of surviving patients represents the percent survival of patients in any of the disclosure categories every 12 months over the study period of 240 months. We evaluated 315 patients, (a, -), patients with negative disclosure regarding AGR2 (-). 100% for 107 patients, {b, -----), borderline revelation patients (+/-) grouped with all other positive disclosure groups (+, ++, +++ / ++++) . 100% for 208 patients.

There were 103 censored observations in a (19 deaths from other causes); and 70 censored observations in b (39 deaths from other causes). The two graphs were very significantly different (Wilcoxon χ2 statistic = 97.40, 1 degree of freedom, P < 0.0001).

EXPERIMENTAL RESULTS EXAMPLE 1: Preparation of an AGR2-specific antibody-producing hybridoma, and production and isolation of the AGR2 antibody 1.1 Sensitizer Antigen The peptide fragment KPGAKKDTKDSRPKL (Sequence ID No. 2), which represents amino acid residues 26 to 40 of AGR2 (as specified in Sequence ID No. 1) was synthesized again using standard synthetic procedures. 1.2 Immunization

Four mice, aged between 6 and 8 weeks were used as host animals to be immunized. The peptide fragment KPGAKKDTKDSRPKL (Sequence ID No. 2) was used as the sensitization antigen. The mice were given 50æg / injection of the KPGAKKDTKDSRPKL sensitization antigen by subcutaneous injection. Injections were administered on days 0, 21 and 42 (with a final booster before fusion) to stimulate an immune response against the antigen and to elicit antibody production. Preimmune serum was collected on day 0. 1.3 Preparation of vim Hybridoma

The best response mouse splenic lymphocytes were fused with Sp2 / 0-Ag-14 myeloma cells using PEG (polyethylene glycol) and the resulting hybridomas applied to 96-well plates for HAT selection. The IgG-producing hybridomas were screened against the antigen. Positive colonies were expanded in assays relative to antibody production and reactivity against the antigen (peptide) and recombinant AGR2 protein. The selected hybridomas were cloned by limiting dilution and screened against the antigen (peptide) and recombinant AGR2 protein. The selected hybridoma was used to define the antibody isotype (Gl kappa) and for IgG production and purification. 1.4 Determination of monoclonal antibody specificity

Figure 1. Western blot of the selected antibody on AGR2 and AGR3. The ability of the monoclonal antibody to specifically bind to AGR2 was determined by western blotting, as follows. The purified recombinant AGR2 (0.5 æg, Lane 1) and recombinant AGR3 (0.5 æg, Lane 2) proteins were subjected to polyacrylamide gel electrophoresis and transfer into PVF membranes. 33

These membranes were then blocked with a solution containing 5% skim milk and incubated with a 1: 2000 dilution of the anti-AGR2 monoclonal antibody overnight.

After incubation overnight, the membrane was washed and incubated for one hour with an HRP-conjugated anti-mouse antibody. The bound antibody was detected by chemiluminescence with an ECL kit and exposure of the filter to photographic film.

The results of this experiment are shown in Figure 1. In this Figure it can be seen that lane 1 shows a strong signal with the AGR2 protein (arrow), but no signal with the AGR3 protein. This result clearly indicates that the monoclonal antibody of the invention, produced according to the protocol outlined above, is specific for AGR2. Immunocytochemical detection of AGR2 with a monoclonal antibody of the invention (produced as described in Example 1), unlike a conventional rabbit polyclonal antibody known in the prior art, enables the discriminatory boundary between tumors with positive and negative development is reduced from 5% to 1% of the revealed carcinoma cells. Although not intended to be limiting in any way, the inventors believe that this increased sensitivity results in the absence of cross-reactivity of the monoclonal antibody with unrelated products, for example AGR3. The immunohistochemical method of tissue sections with anti-AGR2 / PZ7A10F10, the monoclonal antibody of the invention, is carried out as follows. Histological sections (4 pm cuts) on APES coated slides (by a modification of the method of Maddox, P.H. and Jenkins, D.) were degreased in xylene and rehydrated through ethanol to water as previously described by Warburton et al. (1982). Antigen retrieval was performed by microwave (Cuevas et al., 1994). Sections were immersed in 10 mM citrate buffer, pH 6.0 and subjected to microwave at 850 watts for 15 minutes using a domestic microwave oven. The coverslips were allowed to cool for an additional 15 minutes while remaining immersed in the citrate buffer. After recovery of the antigen, the endogenous peroxidase activity in the tissue sections was blocked by immersing the slides in 100% methanol containing 0.05% (v / v) H2 O2 for 20 min at room temperature (Streefkerk, 1972). Anti-AGR2 / PZ7A10F10 monoclonal antibody was applied to the slides at a dilution of 1: 100 in 0.5% BSA / PBS overnight at room temperature in a humidity chamber. Indirect immunocytochemical development was performed using a commercially available improved HRP-labeled polymer system, DAKO EnVision + System, peroxidase (DAB) (Dako Ltd, Ely, UK) (Heras, 1995), prepared according to the manufacturer's instructions. Sections were then washed in running water before being exposed to contrast on Mayer's Hemalum. Those were then dehydrated through graded ethanol and xylene and mounted on DPX fixative (Merck, Poole, UK).

Using this discriminatory levels (made available by the antibody of the invention and assays of the invention), the association of patient survival over time was significantly different between positively and negatively revealed tumors. The significance value χ2 (χ2) (Wilcoxon's statistic) and the relative risk (RR) (univariate Cox analysis) were χ2 = 46.5, RR = 3.8 (95% CI 2.7-5.3 ) using the polyclonal antibody and χ2 = 97.4, RR = 30.5 (95% CI 11-86) using the monoclonal.

This much higher association of IgR2-positive immuno-cytokine revelation with early death of the patient obtained with the monoclonal antibody means that, when compared to other conventional markers (tumor size, histological index, tumor involved lymph nodes), and novel markers (for example, disclosure relative to S100A4, S100P, osteopontin, c-met, ERa, PgR, p53, cathepsin D, c-erbB-2), became the most significant independent prognostic variable in a multivariate regression analysis of Cox (χ2 = 14.3, P < 0.001, RR = 10; 95% CI 3-33).

Previously, the contribution to prognosis of the conventional and novel combined markers meant that the disclosure regarding AGR2 using a polyclonal antibody was not independently associated with survival, and its contribution was balanced by these other markers. Thus, the monoclonal antibodies of the invention, and assays of the invention, provide notable advantages over the prior art. EXAMPLE 3: Diagnostic use of anti-AGR2 antibody - second study

In an expansion of the study reported above, breast cancer tissues from 320 patients suffering from advanced breast cancer and treated by mastectomy and radical mastectomy were investigated using the antibodies of the invention. This study sought to further investigate the usefulness of the antibodies of the invention in prognosis and diagnosis.

Patients from whom the samples were collected were from 1976 to 1982 in general surgery clinics in the Merseyside Region of North West England. This group of patients represented a reasonable image of the population as free medical treatment is available in the United Kingdom. The age range of patients was 29-92 with an average age of 57 years. These patients were followed for an average period of 16 years with a range of 14-20 years. Patient survival was updated on August 31, 1995. The study further investigated the relationships between labeling of samples using the antibodies of the invention, and patient survival rates. Immunological labeling using the anti-AGR2 / 7A10 monoclonal antibody (produced by the hybridoma PZ7A10F10) was performed using the protocols described in the foregoing study (briefly, immunological labeling of degreased histological sections, the bound primary antibody visualized by the Envision system in which the production of a brown color indicates antibody binding).

Revelation analysis

Brown spots were observed when AGR2 was present. The percentage of cancer cells revealed was determined using light microscopy and this value was used to assign the sample to one of six groups, as shown in Table 1 below. Any minor immunological labeling of normal (non-tumor) cells was ignored for the purposes of the study. 37

Statistical Analysis Survival analysis was performed to determine any associations between survival times in each disclosure category in the total follow-up period of 240 months. Only those patients who died of cancer were analyzed. Survival curves were constructed from the life table using Kaplan-Meier plots and survival data were analyzed using Wilcoxon's Statistics (Gehan). Patients who died of causes other than cancer and who were dead at the end of the follow-up period were excluded from the analysis. Univariate Cox regression analysis was performed to obtain relative risk values with 95% confidence intervals for each of the categories being compared. Multivariate Cox regression analysis with all variables present was also performed to determine the relative significance between prognostic factors and to investigate whether the survival of the patient with AGR2 was independent of these prognostic factors. A stepwise selection procedure was used. Cross Classification was used to compare groups of patients separated into the negative (unrevealed) and positive (revealed) categories for two different prognostic markers. Bilateral Fisher Exact Test was performed for any significant difference between the tumors immunocytochemically revealed against AGR2 and those positive for pathological markers or for selected immunohistochemical markers (which were also believed to be of relevance in prognosis or diagnosis). The pathological markers used included histological index, tumor size, and ganglia status while the immunocytochemical marker group included S100P, osteopontin, estrogen receptor α (ERα), c-erbB-2, c-erbB-3, S100A4, progesterone receptor (PgR), p383, cathepsin D, pS2 and c-met in the primary tumor. Data generated from AGR2 labeling using a monoclonal antibody of the invention, and the data generated in the previous study using polyclonal antibody as a comparison, were evaluated for correlation using the kappa function in the Cross Classification. The kappa index was given to show whether there were possible concordances beyond chance between two variables. All statistical calculations were performed using the Statistical Package for Social Sciences version 13.0 (SPSS Inc., Chicago, IL, USA).

RESULTS

Immunocytochemical revelation of AGR2 using monoclonal antibody Most normal breasts were not revealed by the monoclonal antibody against AGR2 (Figure 2A) although there were very rare weak revelations of terminal ductal lobe units (Figure 2B). This disclosure was readily distinguishable from the revelation observed in positively-labeled cancer cells. The development of carcinomas (Figure 2C) was mainly localized in the cytoplasm which had a granular and membrane appearance (Figure 2D).

Carcinoma samples were assessed for development. Patients were divided into 6 categories according to the percentage of carcinoma cells bound by the antibody of the invention (in agreement with Table D). The + + + + group (more than 75% of carcinoma cell development) was pooled with (50-75% of development) due to the relatively small number of samples in these two categories. Of the 320 malignant samples evaluated, 110 (34%) do not contain cells binding the antibody of the invention (samples " no revealed ", shown as -), 94 (30%) exhibited binding of the antibody of the invention to less than 5% of the tumor cells present (" boundary disclosure ", shown as +/-), 62 (19%) exhibited antibody binding to 5-25% of the tumor cells present (shown as +), 26 (8%) exhibited binding of the antibody of the invention to 25-50% of the tumor cells present (shown as ++) and 28 (9%). exhibited binding of the antibody of the invention to 50-100% of tumor cells are present (shown as +++) Figure 3 shows the frequencies of each disclosure category.

Association of AGR2 with other pathological and histological variables Immunocytochemical revelation in relation to AGR2 using monoclonal antibody was evaluated using Cross Classification to compare it with other pathological and immunocytochemical variables. Pathological variables included histological index, tumor size, and ganglia status whereas immunocytochemical variables included the disclosure of AGR2 using a polyclonal antibody, relative to S100P, osteopontin, estrogen receptor α (ERα), c-erbB- 2, c-erbB-3, S100A4, progesterone receptor (PgR), p53, cathepsin D, pS2 and c-met. Comparisons were made using frames two by two and were evaluated using Fisher's Exact Test (bilateral). Immunocytochemical development with respect to AGR2 was adjusted to a threshold of 1%, i.e., the border (+/-) group was incorporated into the other positive developing groups, leaving the negative developing group 40 as those tumors with less than 1% of carcinoma cells revealed. Table 2 shows the association of the revelation with respect to AGR2 with other pathological variables. Only the state of the ganglia (P = 0.036) showed a significant association with the positive revelation regarding AGR2. The comparison was also made at the 5% threshold level of AGR2 (not shown), that is, the boundary group (+/-) was incorporated into the negative (-) development group, leaving the positive development group to consist of all definitely positive carcinomas (+, ++, +++ / ++++). The results were similar; however, the status of the ganglia now showed only a limiting significance (P = 0.056) of association with the disclosure in relation to AGR2. The Cross Classification of the revelation in relation to AGR2 was initially performed at a threshold level of 1%. In contrast, the disclosure to other tumor markers that exhibited an association with the patient's outcome in breast cancer was set at the level that achieved the most significant correlation, generally 5%. Of all molecular variables, the development in relation to S100P (P <0.0001), osteopontin (P <0.001), ERa (P = 0.002), c-erbB-3 (P = 0.001) (P &lt; 0.0001), pS2 (P = 0.004) and c-met (P &lt; 0.0001) showed a significant association with a positive development for AGR2.

If the threshold level for disclosure in relation to AGR2 was set at 5%, the significance of the results obtained was lower than the threshold level of 1% (not shown). This indicates the surprisingly strong prognostic or diagnostic relevance of any binding of the antibody of the invention to tumor cells in a sample of the patient. 41 The significance of AGR2 development related to patient survival

This second study was designed to investigate whether binding of an antibody of the invention to a patient sample (such as tumor cells in a primary tumor sample) is associated with the duration of patient survival in breast cancer. A reduced survival duration is generally associated with metastatic disease resulting from primary tumors. To assess the association, survival curves for patients presenting different levels of antibody binding to tumor cells in their samples were plotted and compared (Figure 4).

Using Wilcoxon Gehan Statistic, there was a significant overall difference in survival times of patients with the 5 different categories of carcinomas revealed (Wilcoxon χ2 test = 103.16, 4 degrees of freedom, P <0.0001). 96% of the 107 patients who were classified as negatively (-) were alive at the end of the study with a median survival time of &gt; 216 months. In the frontier developing group (+/-), the cumulative proportion of surviving patients was 38% for 93 patients and the median survival time was 8 7 months. For those patients with tumors that had positive developing 5-25% (+) using the antibody of the invention, the cumulative surviving ratio was 21% and the median survival time was 65 months. There were no survivors (0%) among the 26 patients in the tumor group containing 25-50% (+) of carcinoma cells binding the antibody of the invention, and the median survival time was only 45 months. For those patients whose tumors contained more than 50% carcinoma cells (+++ / ++++) binding an antibody of the invention, the cumulative ratio surviving for 28 patients was 21% and the median survival time was 52 months (Figure 4). The relative risk (R.R.) and its 95% confidence interval (95% C.I.) corresponding to different associations of disclosure category are detailed in Table 4.

In Figure 6, of the 107 patients whose tumor cells were classified as not binding an antibody of the invention, 96% of them survived until the end of the study with a median survival time &gt; 216 months. The cumulative survival rate for those who showed positive development for AGR2 (+ / -, +, ++, +++ / ++++) was 26% at the end of the study with a median survival time of 69 months . The R.R. was high at 30.5 with 95% CI, 11.3-82.5 (Figure 6). This illustrates that any binding of an antibody of the invention to tumor cells present in a patient sample is prognostic and / or diagnostic of metastatic disease. That this is a more accurate and sensitive marker than binding using other anti-AGR2 antibodies known in the prior art is clearly indicated by the results of the foregoing study (in which the usefulness of the antibodies according to the invention in diagnosis / prognosis was compared with that of known polyclonal antibodies raised against AGR2).

Determination of whether the survival of the patient with AGR2 was independent of other prognostic factors

Survival data for samples containing tumor cells that bind to the antibodies of the invention and comparable data from all pathological variables and most of the tumor variables that were previously assessed using Cross Classification were introduced in a Cox multiple regression analysis (Materials and Methods). The disclosure for AGR2 was set at the optimum threshold level of 1% while the disclosure for most other parameters was set at the optimum threshold level of 5% for patient survival in the univariate analysis. The revelations regarding c-erbB-3 and pS2 were omitted from the analysis, since they did not show a significant correlation with patient survival in the univariate analysis in this particular group of patients. Cathepsin D was fixed at a threshold level of 1% since only this threshold showed significance in the univariate analysis in an earlier study with this particular group of patients. Of all the factors investigated, only binding of an antibody of the invention, or disclosure using osteopontin, c-erbB-2, ERa, c-met and S100A4 were independent and correlated significantly with patient survival times. The presence of cells binding to an antibody of the invention (shown as &quot; AGR2 &quot;) was found to be the most significant independent variable in relation to patient survival in this analysis (Table 5).

Sequence Information

Sequence ID No. 1 (Human AGR2 amino acid sequence) mekipvsafl llvalsytla rdttvkpgak kdtkdsrpkl pqtlsrgwgd qliwtqtyee alyksktsnk plmiihhlde cphsqalkkv faenkeiqkl aeqfvllnlv yettdkhlsp dgqyvprimf vdpsltvrad itgrysnrly ayepadtall ldnmkkalkl lktel

Sequence ID No. 2 (Amino acid sequence of the sensitizing antigen) kpgakkdtkdsrpkl 44

Frames

Percentage of tumor cells revealed (%) Classification <1 - 1-5 +/- 5-25 + 25-50 + + 50-75 + + 75-100 + + + +

Caption Table 1. Classification of breast cancer patients based on primary tumor immunological labeling using antibodies of the invention. The system used was based on the percentage of malignant cells that scored positively using an antibody of the invention.

Table 2. Association of the revelation with respect to AGR2 at the threshold level of 1% with pathological variables.

Fisher's exact test Number (%) of Number (%) of AGR2â AGR2a

(Category) (2) (1) (2) (3) (3) (3) (3) (3) , 5) Tumor (total 304) T1, 2 83 (80.6) 150 (74.6) T3.4 20 (19.4) 51 (25.4) 0.256 Ganglion (total 231) Ganglion (-) 48 63.2) 75 (48.4) 0.036 Ganglion (+) 28 (36.8) 80 (51.6)

Table 2 Legend of immunocytochemistry with respect to AGR2 at the threshold level of 1%. (-) indicates the negative group while (+) indicates all other positive groups including the frontier disclosure group (+/-). b Fisher's Exact Test (bilateral) was used to evaluate the significance of any association between AGR2 and other pathological variables. Those that were significant (P &lt; 0.05) are shown in bold. Table 3. Association of the disclosure in relation to AGR2 at the threshold level of 1% with the disclosure in relation to molecular markers.

Dissolution of molecular variables (number of patients) S100P (total 280) S100P (-) S100P (+) Osteopontin (total 306) OPN (-) OPN (+) ERa (total 313) ERa (-) ERa (+) C (+) C-erbB-3 (+) C-erbB-3 (+) C-erbB-2 (+ 319) c-erbB- (+) Cat. D (-) Cat D (-) Cat. D (-) Cat. D (-) Cat. D (+) Number (%) Number (%) Accurate AGR2b Test of AGR2b Fisherc (+) (bilateral) 82 (82.0) 18 (18.0) 74 (68.5) 34 (31, 5) 64 (58.7) 45 (41.3) 87 (80.6) 21 (19.4) 58 (53.7) 50 (46.3) 96 (87.3) 14 (12.7) 79 (75.2) 26 (24.8) 71 (65.1) 38 (34.9) 38 (47.5) 42 (52.5) 51 (28.3) 129 (71.7) 29 14.6) 169 (85.4) 81 (39.7) 123 (60.3) 155 (75.2) 51 (24.8) 67 (33.3) 134 (66.7) 96 (46, 2) 112 (53.8) 109 (54.5) 91 (45.5) 122 (58.4) 87 (41.6) <0.0001 <0.0001 0.002 0.324 0.001 <0 , 0001 &lt; 0.0001 0.277 66 (39.1) 0.218 103 (60.9) 46 46 Disclosure of molecular variables3 (No. of patients) N (+) (bilateral) pS2 (total 315) pS2 (-) 76 (69.7) 108 (52.4) 0.004 pS2 (+) 33 (30.3) 98 (47.6) 56 (57.7) 41 (42.3) 28 (15.3) 155 (84.7) <0.0001 C-met (total 280 ) c-met (-) c-met (+)

Legend of Table 3 3 Immunocytochemical revelation of molecular variables at the threshold level of 5%. - and + / - are grouped as (-) and +, + +, + + + + + + + + are grouped as (+). b Immunocytochemical revelation in relation to AGR2 at the threshold level of 1%. (-) indicates the negative group and (+) indicates all other groups (+ / -, +, + +, + + + / + + + +). c Fisher's Exact Test (bilateral) was used to evaluate any significant association between the revelation in relation to AGR2 and the revelation in relation to other molecular variables. Those that were significant (P &lt; 0.005) are shown in bold.

Table 4. Comparison paired between each disclosure category in relation to AGR2.

Development Categories 3 x ^ b df c P d V <D 95% CI and Global 103.16 4 <0.0001 NANA - e + / - 68.19 1 <0.0001 24.178,71 -66.6 - and + 84.27 1 <0.0001 33.9 12.1-94.8 + + CD 1884, 80 1 <0.0001 52.9 18.1-155 - e + + + 60.18 1 <0.0001 36.1 12.2-107 +/- and + 1, 08 1 0.298 1, 41 0.94-2.10 +/- and ++ 4.53 1 0.033 2.20 1.33-3.63 +/- and +++ 1.71 1 0.192 1.50 0.88-2.56 + and + + 2.36 1 0, 124 1.56 0.93- 2.61 + and + + + 0.656 1 0, 418 1.07 0.62-1.84 ++ and +++ 0.026 1 0.871 1.46 0.78-2.73 47

Categories of χΤΈ d. The developmental categories relative to AGR2 are defined in Table 1 and the paired comparisons correspond to those in the survival charts of Figure 5. b χ2 test using Wilcoxon's statistics . c Degrees of freedom. d Probability values of χ2. P values that were significant are shown in bold. and relative patient survival risk (R.R.) and 95% confidence interval (95% C.I.) of the Cox multiple regression analysis.

Table 5. Results of Cox multiple regression analysis using all pathological variables and most of the molecular variables.

Tumor Variables a β * SE c d.f. P and RR f 95% CI f mAGR2 2.30 0.61 14.3 1 <0.0001 10.0 3.03-33.1 Osteopontin 1.65 0.61 7.31 1 0.007 5.223 , 58-17.3 c-erbB-2 0.82 0.31 6, 83 1 0.009 2.27 1.23-4.21 ERa -0.65 0.27 5, 59 1 0.018 0.52 0, 31-0.90 c-met 1.61 0.76 4, 52 1 0.034 5.00 1.13-22.0 S100A4 0.54 0.31 2.97 1 0.085 1.71 0.93-3, 16

Legend of Table 5 a Tumor variables that showed a significant and independent association with patient survival in the multivariate analysis. The variables that did not reach a significant independent level with patient survival were S100P, histological index, tumor size, state of the ganglia, p53, PgR, Cathepsin D. b Parameter value β in the Cox multiple regression analysis c Error standard of β. d χ2 statistic generated from the Cox multiple regression analysis and χ2 probability. Global χ2 = 77.83, 7 degrees of freedom, P &lt; 0.0001. Relative risk (R) for patient survival and 95% confidence interval (95% CI) of the multivariate analysis. 48

REFERENCES

Innes H.E., Liu D., Barraclough R., Davies M.P.A., O'Neill P.A., Platt-Higgins A., Silva Rudland S., Sibson D.R. and Rudland P.S. (2006). Significance of the metastasis-inducing protein AGR2 for outcome in hormonally-treated breast cancer patients. Br. J. Cancer 94, 1057-1065

Maddox, P.H. and Jenkins, D., 3-Aminopropyltriethoxysilane (APES): a new advance in section adhesion. Clin Pathol. 1987; 40: 1256-1257

Warburton, M.J., Mitchell, D., Ormerod, E.J., and Rudland, P.S. Distribution of myoepithelial cells and basement membrane proteins in the resting, pregnant, lactating and involuting rat mammary gland. (1982) J. Histochem. Cytochem., 30: 667-676

Cuevas, E.C., Bateman, A.C., Wilkins B.S. et al. Microwave antigen retrieval in immunocytochemistry: a study of 80 antibodies, (1994) J. Clin. Pathol., 47, 448-452

Streefkerk, J.G. Inhibition of erythrocyte pseudoperoxidase activity by treatment with hydrogen peroxide following methanol. (1972) J. Histochem. Cytochem., 20, 829

Heras, A. et al. Enhanced label-polymer system for immunohistochemistry. XVth Eur Cong Pathol. Copenhagen, Denmark 1995; Sept 3-8

Fletcher, G. et al. hAG-2 and hAG-3, human homologues of genes involved in differentiation, are associated with estrogen receptor-positive breast tumors and interact with metastasis gene C4-4a and dystroglycan (2003). Br-J. Cancer, 88, 579-585. 49

SEQUENCE LISTING

&lt; 110 &gt; THE UNIVERSITY OF LIVERPOOL BARRACLOUGH, ROGER BARRACLOUGH, DONG LIU RUDLAND, PHILIP

&lt; 120 &gt; ANTIBODIES, TESTS and HYBRIDOMAS

&lt; 130 &gt; P90181PWO &lt; 150 &gt; GB0616929.6 &lt; 151 &gt; 2006-08-26 &lt; 160 &gt; 2 &lt; 170 &gt; Patentln version 3.3

&lt; 210 &gt; 1 &lt; 211 &gt; 175 &lt; 212 &gt; PRT &lt; 213 &gt; Homo sapiens &lt; 400 &gt; 1

Met Glu Lys lie Pro Will Be Ala Phe Leu Leu Leu vai Ala Leu Ser 15 10 15

Tyr Thr Leu Ala Arg Asp Thr Thr goes Lys Pro Gly Ala Lys Lys Asp 20 25 30

Thr Lys Asp be Arg Pro Lys Leu Pro Gin Thr Leu be Arg Gly Trp 35 40 45

Gly Asp Gin Leu ile Trp Thr Gin Thr Tyr Glu Glu Ala Leu Tyr Lys 50 55 60 being Lys Thr Ser Asn Lys Pro Leu Met le ile His His Leu Asp Glu 65 70 75 80 cys Pro His Ser Gin Ala Leu Lys Lys Le Phe Ala Glu Asn Lys Glu 85 90 95 50 Ile Gin Lys Leu Ala Glu Gin Phe Leu Leu Asn Leu Val Tyr Glu 100 105 110 Thr Thr Asp 115 Lys His Leu Ser Pro 120 Asp Gly Gin Tyr Val 125 Pro Arg Ile Met Phe Asp pro Ser Leu Thr val Arg Ala Asp ile Thr Gly Arg 130 135 140 Tyr Ser Asn Arg Leu Tyr Ala Tyr Glu Pro Ala Asp Thr Ala Leu Leu 145 150 155 160 Leu Asp Asn Met Lys Lys Ala Leu Lys Leu Leu Lys Thr Glu Leu 165 170 175

&lt; 210 &gt; 2 &lt; 211 &gt; 15 &lt; 212 &gt; PRT &lt; 213 &gt; Homo sapiens &lt; 400 &gt; 2

Lys ργο Gly Ala Lys Lys Asp Thr Lys Asp being Arg Pro Lys Leu 15 10 15

Lisbon, October 19, 2010

Claims (17)

A monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2. The monoclonal antibody, or antigen-binding fragment of the one of claim 1, which does not bind to AGR3; and / or binding to the same antigen as the monoclonal antibody produced by the hybridoma PZ7A10F10 deposited under the Budapest Treaty and to which accession number 06082201 has been given. A monoclonal antibody, or antigen-binding fragment thereof, according to claim 1 or claim 2, wherein the antibody or fragment is labeled with a reporter unit selected from the group consisting of; a luminescent unit; a bioluminescent unit; a radioactive material; a prosthetic group; a colorimetric unit; detectable nanoparticles; and a chromogenic unit; and a fluorescent moiety, which may be selected from the group consisting of: fluorescein isothiocyanate (FITC); rhodamine (TRITC); phycoerythrin; allophycocyanin; coumarin (AMCA); Texas red; cyanine (Cy 2), cyanine (Cy 3); and cyanine (Cy5). An in vitro method for diagnosing cancer using an antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2). The method of diagnosing cancer according to claim 4, wherein the method comprises: contacting a patient sample with a monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to a epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2); and analyzing for binding of the antibody to the patient sample, wherein binding of the antibody to the patient sample is diagnostic of metastatic disease in the patient. An in vitro method for evaluating the likelihood of developing cancer using an antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2). A method for evaluating the likelihood of developing cancer according to claim 6, wherein the method comprises: contacting a patient sample with a monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2); and analyze for binding of the antibody to the patient sample, wherein binding of the antibody to the patient sample indicates that the patient has a high probability of developing metastatic disease. A method according to any one of claims 7 to 10, wherein the cancer is a metastatic cancer. A method according to any one of claims 48, wherein the analysis of antibody binding to the patient sample utilizes a technique selected from the group consisting of: immunocytochemical labeling of a tissue sample; an enzyme-linked immunosorbent assay (ELISA); chemiluminescent immunoassay; fluorescence immunoassay; and fluorescence-activated cell sorting (FACS). A method according to any one of claims 4 to 9, wherein: i) the patient sample comprises tumor cells, and the assay comprises analyzing the binding of the antibody, or antigen-binding fragment thereof, to tumor cells in patient sample; and / or ii) the patient sample is selected from the group consisting of: a blood sample; a biopsy sample; a histological sample; and a cryotomy sample; and / or the antibody, or antigen-binding fragment thereof, used is as defined in any one of claims 1 to 3. A kit comprising a monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2). A kit according to claim 11 comprising a monoclonal antibody, or antigen-binding fragment thereof, produced by the hybridoma PZ7A10F10 deposited under the Budapest Treaty and to which accession number 06082201 has been given; and / or at least one item selected from the group consisting of: instructional material on how to use the kit; reagents for use in detecting antibody binding; reagents to be used in antibody incubation; and agents for visualization of cell nuclei. A solid support comprising an antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) immobilized thereon. A solid support according to claim 13 comprising a monoclonal antibody, or antigen-binding fragment thereof, produced by the hybridoma PZ7A10F10 deposited under the Budapest Treaty and to which accession number 06082201 has been given; and / or wherein the solid support is a matrix. 15. Hybridoma PZ7A10F10 deposited under the Budapest Treaty and to which Accession Number 06082201 has been given. A monoclonal antibody, or an antigen-binding fragment thereof, which specifically binds to an epitope within the sequence KPGAKKDTKDSRPKL (Sequence ID No. 2) of AGR2 to be used as a medicament. A monoclonal antibody, or an antigen-binding fragment thereof, according to claim 16 for use as a medicament in the prevention and / or treatment of cancer and / or an inflammatory disease. Lisbon, October 19, 2010